阅读说明：本技术 一种基于广义短路比法的风电并网低频振荡抑制方法 (Wind power grid-connected low-frequency oscillation suppression method based on generalized short-circuit ratio method ) 是由 张飞云 卜京 孙莹 卞婉春 郑铭洲 夏凡吴双 殷明慧 谢云云 邹云 刘建坤 周前 于 2019-08-16 设计创作，主要内容包括：本发明公开了一种基于广义短路比法的风电并网低频振荡抑制方法。该方法为：首先建立风电并网中的网侧模型,并根据直流电容模型,用广义短路比法简化线路；将电容电压用正弦函数表示,通过扫频法得到能够激发网侧低频振荡模态的直流电容振荡频率；然后建立风电并网机侧模型,得到直流电容关于风机转速的传递函数；接着建立风速对风机转速影响模型,分析得到能够激发直流电容振荡频率的风速波动频段；最后利用粒子群算法协调优化机侧和网侧控制器参数,使得相应模态阻尼比最大,从而达到抑制低频振荡的效果。本发明具有结构简单、硬件成本低、可靠性高的优点。(The invention discloses a wind power grid-connected low-frequency oscillation suppression method based on a generalized short-circuit ratio method. The method comprises the following steps: firstly, establishing a grid side model in wind power integration, and simplifying a circuit by using a generalized short circuit ratio method according to a direct current capacitance model; the capacitor voltage is expressed by a sine function, and the direct current capacitor oscillation frequency capable of exciting the low-frequency oscillation mode of the network side is obtained by a frequency sweeping method; then establishing a side model of the wind power grid-connected machine to obtain a transfer function of the direct current capacitor about the rotating speed of the fan; then establishing a model of influence of wind speed on the rotating speed of the fan, and analyzing to obtain a wind speed fluctuation frequency band capable of exciting the oscillation frequency of the direct current capacitor; and finally, coordinating and optimizing parameters of the machine side controller and the network side controller by utilizing a particle swarm algorithm to enable the corresponding modal damping ratio to be maximum, thereby achieving the effect of inhibiting low-frequency oscillation. The invention has the advantages of simple structure, low hardware cost and high reliability.)

1. A wind power grid-connected low-frequency oscillation suppression method based on a generalized short-circuit ratio method is characterized by comprising the following steps:

step 1, establishing a grid side model in wind power integration, and simplifying a circuit by using a generalized short circuit ratio method according to a direct current capacitance model;

step 2, expressing the capacitor voltage by a sine function, and obtaining the direct current capacitor oscillation frequency capable of exciting a network side low-frequency oscillation mode by a frequency sweeping method;

step 3, establishing a wind power grid-connected machine side model to obtain a transfer function of the direct current capacitor about the rotating speed of the fan;

step 4, establishing a model of influence of wind speed on the rotating speed of the fan, and analyzing to obtain a wind speed fluctuation frequency band capable of exciting the direct current capacitance oscillation frequency in the step 2;

and 5, coordinating and optimizing parameters of the machine side controller and the network side controller by utilizing a particle swarm algorithm to enable the corresponding modal damping ratio to be maximum, so that low-frequency oscillation is suppressed.

2. The wind power grid-connected low-frequency oscillation suppression method based on the generalized short-circuit ratio method according to claim 1, characterized in that the step 1 of establishing a grid-side model in wind power grid connection and simplifying lines by the generalized short-circuit ratio method according to a direct-current capacitance model specifically comprises the following steps:

step 1.1, dividing a network side model in wind power integration into Jacobian transfer function matrix J of a power electronic device group_Gm(s) and Jacobian transfer function matrix J for an AC power grid_netm(s)：

Wherein, Δ P represents the fluctuation value of the active power of the network side, Δ Q represents the fluctuation value of the reactive power of the network side, Δ θ represents the fluctuation value of the phase angle of the voltage of the network side, Δ U represents the fluctuation value of the voltage of the network side, and U represents the amplitude of the voltage of the network side;

step 1.2, Jacobian transfer function matrix J of alternating current power grid_netm(s) is:

wherein, J_Pθ(s) a transfer function representing the relationship between the active power of the AC network and the phase angle of the network-side voltage, J_PU(s) a transfer function representing the relationship between the active power of the AC network and the network-side voltage amplitude, J_Qθ(s) a transfer function representing the relationship between the reactive power of the AC network and the phase angle of the network-side voltage, J_QU(s) a transfer function representing the relation between the reactive power of the ac power grid and the grid side voltage amplitude;

in the formula (I), the compound is shown in the specification,

wherein α(s) ═ 1/[ (s/ω)₀)²+1]，β(s)＝(s/ω₀)/[(s/ω₀)²+1]Wherein ω is₀The power frequency angular velocity; p_i＝diag(PL₁,PL₂,…,PL_n)，P_iIs a per unit value, PL, of the output power of the power electronic equipment based on unit capacity₁,PL₂,…,PL_nIndicating the power of the grid-connected line injection node 1, 2, …, n; m is a cosine value of the current injected between the nodes i and j; n is the sine value of the injected current between the nodes i and j;

wherein:

in the formula of U_i、U_jRepresents the node voltages of the node i and the node j; theta_ijRepresents the voltage phase angle difference between node i and node j;

B_ijin order to eliminate elements in the node admittance matrix B after the nodes connected with the infinite power grid, the specific expression is as follows:

in the formula, L_ijRepresenting the inductance of the transmission line between node i and node j; n is the total number of nodes contained in the line;

step 1.3, setting the Jacobian transfer function matrixes of all the power electronic equipment to be the same under the self-reference capacity, converting the Jacobian transfer function matrixes of all the power electronic equipment to be under the system capacity, and obtaining a dynamic Jacobian transfer function matrix J comprising all the power electronic equipment_Gm(s) and the corresponding transfer functions are:

in the formula, S_B＝diag(S_B1,S_B2,…,S_Bn)，S_BiCapacity of the ith equipment;

the set wind speed is constant, namely the output of the fan is constant, and the direct current capacitance link is as follows:

in the formula,. DELTA.V_dcRepresenting the fluctuation quantity of voltage at two ends of the direct current capacitor; delta P_gRepresenting the fluctuation quantity of the active power at the network side; c represents a dc capacitance value; v_dc0Representing the steady-state value of the voltage at two ends of the direct current capacitor;

substitution of formula (7) into J_GmIn the step(s), the dynamic jacobian transfer function of the power electronic equipment comprising the direct current capacitance link is obtained as follows:

in the formula, G₁(s) is the transfer function of the power outer loop control loop; g₂(s) is the current inner loop control loop transfer function; g_PLL(s) is a transfer function of the phase-locked loop control loop; l is the filter inductance; i is_dIs the net side d-axis current; i is_qIs the net side q-axis current; u shape_dIs a grid-connected voltage;

step 1.4, mixing J_netm(s) and J_GmThe concrete form of(s) is substituted for the closed-loop characteristic equation of the multi-feed system of formula (1) as follows:

in the formula, det (×) represents a determinant;

simultaneous left multiplication matrix of terms in equation (9)

in the formula, P_b＝P_i/S_Bi，Q_b＝Q_i/S_Bi；I_n∈R^n×nIs an identity matrix; j. the design is a square_eqTo expand the admittance matrix, the expression is:

extended admittance matrix J_eqThe eigenvalue decomposition can be carried out, all eigenvalues are positive numbers, and the algebraic multiplicity and geometric multiplicity of the minimum eigenvalue are both 1, so that the existence of the invertible matrix W can be applied to J_eqCarrying out similar diagonal transformation to satisfy

W^-1J_eqW＝Λ＝-diag(λ_i) (12)

In the formula, Λ represents J_eqIs a characteristic value of_iA diagonal matrix arranged in order; diag denotes determinant;

step 1.5, because G in formula (10)_Pθ(s)I_n、G_PU(s)I_n、G_Qθ(s)I_n、G_QU(s)I_n、diag(P_b) And diag (Q)_b) All are diagonal matrices, so formula (12) is substituted for formula (10):

at this time, each term in the system closed-loop characteristic equation (8) is a diagonal matrix, so equation (8) is rewritten as follows:

in the formula (I), the compound is shown in the specification,

namely, it is

c₁(s)×c₂(s)…×c_n(s)＝0 (15)

In the formula, c_i(s) a closed loop equation representing the ith branch of the system;

wherein

Each factor c in the closed-loop characteristic equation of the multi-feed system_iAnd(s) has the same form as the closed-loop characteristic equation of the single-feed system, so that the closed-loop characteristic equation of the n-feed system is the product of the closed-loop characteristic equations of the n single-feed systems.

3. The wind power grid-connected low-frequency oscillation suppression method based on the generalized short-circuit ratio method according to claim 2, wherein the capacitor voltage is expressed by a sine function in the step 2, and the direct-current capacitor oscillation frequency capable of exciting the grid-side low-frequency oscillation mode is obtained by a frequency sweep method, and specifically the method comprises the following steps:

step 2.1, mixing V_dcSince the transfer function is represented by the frequency domain, the fluctuation of (2) is represented by a sine function, which is subjected to a laplacian transform to obtain:

in the formula, ω₁Represents the oscillation frequency of the direct current capacitor;

step 2.2, substituting formula (17) into J_GmIn(s), the jacobian transfer function for obtaining the dynamic state of the power electronic device including the capacitance fluctuation is as follows:

step 2.3, comparing the characteristic roots without fluctuation and with fluctuation by a frequency sweeping method to obtain the direct current capacitance oscillation frequency omega capable of exciting the network side low-frequency oscillation mode₁。

4. The wind power grid-connected low-frequency oscillation suppression method based on the generalized short-circuit ratio method according to claim 3, wherein the wind power grid-connected machine side model is established in the step 3 to obtain a transfer function of the direct-current capacitor with respect to the rotating speed of the fan, and the method specifically comprises the following steps:

step 3.1, setting the power of the network side to be free of fluctuation, and establishing a direct current capacitance model:

in the formula,. DELTA.V_dcRepresenting voltage fluctuation values on two sides of the direct current capacitor; delta P_wRepresenting the machine side active power fluctuation value;

step 3.2, establishing a machine side model, wherein the Jacobian transfer function matrix from the fan to the direct current capacitor is as follows:

in the formula, v_ds0、v_qs0Steady-state voltages of a d axis and a q axis of the stator are respectively; i.e. i_ds0、i_qs0The steady-state currents are respectively a d axis and a q axis of the stator; j. the design is a square_MIs the rotational inertia of the fan; p is the number of pole pairs of the fan; psi is the flux linkage value of the permanent magnet fan; delta omega is a machine side rotating speed fluctuation value; a represents machine side active power fluctuation and i_ds0A value of interest; b represents machine side active power fluctuation and v_ds0A value of interest; c represents machine side active power fluctuation and i_qs0A value of interest;

and 3.3, substituting the direct current capacitor model into a formula (20) to obtain a transfer function of the rotating speed to the voltage at two ends of the direct current capacitor, wherein the transfer function is as follows:

5. the wind power grid-connected low-frequency oscillation suppression method based on the generalized short-circuit ratio method according to claim 1, 2, 3 or 4, wherein the wind speed influence model on the rotating speed of the fan is established in the step 4, and a wind speed fluctuation frequency band capable of exciting the direct current capacitance oscillation frequency in the step 2 is obtained through analysis, and specifically the method comprises the following steps:

step 4.1, wind speed v_windFor the input variable, the relation between the output power of the fan and the wind speed is obtained as follows:

wherein ρ is an air density; pi is the circumference ratio; r is the length of the fan blade; v. of_windIs the wind speed; p_mOutputting power for the fan;

and 4.2, expressing the wind speed fluctuation by a sine function:

in the formula, v_wind0Is a steady state value of wind speed; omega_windRepresenting the wind speed oscillation frequency;

the shafting model of the generator related to the wind speed is as follows:

in the formula, J_wRepresenting the generator moment of inertia; t is_mRepresenting generator mechanical torque; t is_eRepresenting the electromagnetic torque of the generator; omega_m0Representing a steady state value of the rotating speed of the generator rotor;

4.3, linearizing the formula (24) to obtain the influence of the wind speed on the rotating speed, and obtaining a wind speed fluctuation frequency band capable of exciting low-frequency oscillation of the direct current capacitor voltage by using a frequency sweeping method;

and 4.4, analyzing the influence of each controller parameter on the oscillation mode to obtain a dominant oscillation mode, and verifying the small signal.

6. The wind power grid-connected low-frequency oscillation suppression method based on the generalized short-circuit ratio method according to claim 5, wherein the parameters of the machine-side and network-side controllers are coordinated and optimized by using the particle swarm algorithm in step 5, so that the corresponding modal damping ratio is maximized, thereby suppressing the low-frequency oscillation, specifically as follows:

step 5.1Controlling the loop proportional parameter PLL by the phase-locked loop_lAnd integral parameter PLL_iThe method comprises the following steps of establishing a multi-controller parameter coordination optimization model by taking a power outer ring proportional parameter kp1, a power outer ring integral parameter ki1, a stator d-axis current inner ring proportional parameter kp2, a stator d-axis current inner ring integral parameter ki2, a stator q-axis current inner ring proportional parameter kp3 and a stator q-axis current inner ring integral parameter ki3 of a machine-side controller, taking a grid-side d-axis current inner ring proportional parameter gp1, a power outer ring integral parameter gi1, a grid-side d-axis current inner ring proportional parameter gp2, a grid-side d-axis current inner ring integral parameter gi2, a grid-side q-axis current inner ring proportional parameter gp3 and a grid-side q-axis current inner ring integral parameter gi3 of a grid-side controller as optimization variables, taking an oscillation mode caused by wind speed fluctuation as a target mode, and taking a target mode damping ratio maximization as an optimization target:

in the formula, η_iDetermining the weight coefficient of the target mode i by a dispatching department according to the actual condition of the power system so as to reflect the attention degree of the dispatching department to different target modes; zeta_iDamping ratio for target mode i; k is the target mode number; f is the maximum system damping ratio of the target mode; max () represents taking the maximum value; Σ denotes a summation function;

step 5.2, in the process of suppressing oscillation by coordinating and optimizing parameters of the network side controller and the machine side controller, the following constraint conditions need to be met: (1) the damping ratio of the oscillation mode caused by wind speed fluctuation needs to be larger than a set threshold value rho₀(ii) a (2) The damping ratio of other oscillation modes which do not need to be optimized is not lower than the set threshold value rho₁Meanwhile, when the oscillation mode damping ratio which does not need to be optimized falls, the falling amplitude needs to be within the lower limit value constraint; (3) the controller parameters to be optimized need to satisfy their own parameter constraints;

then the constraint conditions for obtaining the coordination optimization are as follows:

in the formula, k_p-x,yThe y scale factor of the x controller; k is a radical of_p-x,y-maxAnd k is_p-x,y-minAre each k_p-x,yThe corresponding upper and lower limits; k is a radical of_i-x,yThe ith integral coefficient of the xth controller; k is a radical of_i-x,y-maxAnd k is_i-x,y-minAre each k_i-x,yThe corresponding upper and lower limits; rho_0i1A damping ratio threshold for a given target mode i; rho_0i2ξ as the damping ratio of the target mode i under the initial parameters of the controller_kDamping ratio for non-target mode k, Δ ξ_kThe damping ratio variation before and after coordination optimization for the non-target mode k; rho_1kA damping ratio threshold for a given non-target mode k; rho_2kPercent change in damping ratio for a given non-target mode k, which is a positive number;

and 5.3, obtaining the dominant controller parameters of the target oscillation mode in the step 4.4, solving different dominant controller parameters by using a particle swarm algorithm, obtaining the controller parameter value corresponding to the maximum target function shown in the formula (25), namely the controller parameter value of the mode with the maximum damping ratio, and accordingly inhibiting oscillation of the mode.

Technical Field

The invention belongs to the technical field of power system stabilization and control, and particularly relates to a wind power grid-connected low-frequency oscillation suppression method based on a generalized short-circuit ratio method.

Background

Among the various renewable energy sources, the wind energy is widely distributed, the wind power generation cost is relatively low, and the wind power generation system has great commercial potential and vitality. The use process of the wind energy is clean and pollution-free, and the contradiction between environmental protection and economic development can be well relieved while stable power supply is provided for the modernized construction of the Chinese socialism. In addition, the wind power plant does not occupy fertile land resources on a large scale, and normal production of the planting industry and the farming and animal husbandry is not influenced, and because the areas with rich wind power resources are often in remote wastelands or mountainous regions, the development of wind power energy can drive the development of the tourism industry. With the national strong support for renewable energy power generation in policy, the wind power construction in China also enters a period of rapid development.

Wind power is used as a renewable energy source and has the characteristics of volatility and intermittence, in addition, due to the characteristic of wind power resource distribution, a large-scale wind power generation field is often located in relatively remote areas, the loads of the areas are often relatively small, and a power grid needs to be accessed through a long-distance power transmission line, so that the influence of wind power access on the damping characteristic of a power system is particularly remarkable, a method for suppressing low-frequency oscillation of the system caused by wind speed fluctuation is researched, and the method is a problem which needs to be solved urgently when the wind power grid is connected and the system is analyzed after the large-scale wind power is accessed.

Because a wind turbine grid-connected structure is complex, a traditional small signal analysis method is large in modeling dimension, and no method capable of describing the influence of the frequency of wind speed fluctuation on safe and stable operation of a grid side in detail through a mechanism exists at present, so that the coordination optimization of controller parameters is the joint adjustment of 12 parameters of 6 controllers, and the dimension of the optimization problem is large. Although the existing analysis can simplify the line, the interaction influence of the whole system is not comprehensively considered, and various operation modes of the system and wind speed conditions of a wind power plant cannot be comprehensively considered.

Disclosure of Invention

The invention aims to provide a wind power grid-connected low-frequency oscillation suppression method based on a generalized short-circuit ratio method, which is simple in structure, low in hardware cost and high in reliability.

The technical solution for realizing the purpose of the invention is as follows: a wind power grid-connected low-frequency oscillation suppression method based on a generalized short-circuit ratio method comprises the following steps:

step 1, establishing a grid side model in wind power integration, and simplifying a circuit by using a generalized short circuit ratio method according to a direct current capacitance model;

step 2, expressing the capacitor voltage by a sine function, and obtaining the direct current capacitor oscillation frequency capable of exciting a network side low-frequency oscillation mode by a frequency sweeping method;

step 3, establishing a wind power grid-connected machine side model to obtain a transfer function of the direct current capacitor about the rotating speed of the fan;

step 4, establishing a model of influence of wind speed on the rotating speed of the fan, and analyzing to obtain a wind speed fluctuation frequency band capable of exciting the direct current capacitance oscillation frequency in the step 2;

and 5, coordinating and optimizing parameters of the machine side controller and the network side controller by utilizing a particle swarm algorithm to enable the corresponding modal damping ratio to be maximum, so that low-frequency oscillation is suppressed.

Further, the step 1 of establishing a grid-side model in the wind power grid connection, and simplifying the line by using a generalized short-circuit ratio method according to the direct-current capacitance model specifically includes:

step 1.1, dividing a network side model in wind power integration into Jacobian transfer function matrix J of a power electronic device group_Gm(s) and Jacobian transfer function matrix J for an AC power grid_netm(s)：

Wherein, Δ P represents the fluctuation value of the active power of the network side, Δ Q represents the fluctuation value of the reactive power of the network side, Δ θ represents the fluctuation value of the phase angle of the voltage of the network side, Δ U represents the fluctuation value of the voltage of the network side, and U represents the amplitude of the voltage of the network side;

step 1.2, Jacobian transfer function matrix J of alternating current power grid_netm(s) is:

wherein, J_Pθ(s) a transfer function representing the relationship between the active power of the AC network and the phase angle of the network-side voltage, J_PU(s) a transfer function representing the relationship between the active power of the AC network and the network-side voltage amplitude, J_Qθ(s) a transfer function representing the relationship between the reactive power of the AC network and the phase angle of the network-side voltage, J_QU(s) a transfer function representing the relation between the reactive power of the ac power grid and the grid side voltage amplitude;

in the formula (I), the compound is shown in the specification,

wherein the content of the first and second substances,

wherein ω is₀The power frequency angular velocity; p_i＝diag(PL₁,PL₂,…,PL_n)，P_iIs a per unit value, PL, of the output power of the power electronic equipment based on unit capacity₁,PL₂,…,PL_nIndicating the power of the grid-connected line injection node 1, 2, …, n; m is a cosine value of the current injected between the nodes i and j; n is the sine value of the injected current between the nodes i and j;

wherein:

in the formula of U_i、U_jRepresents the node voltages of the node i and the node j; theta_ijRepresents the voltage phase angle difference between node i and node j;

B_ijin order to eliminate elements in the node admittance matrix B after the nodes connected with the infinite power grid, the specific expression is as follows:

in the formula, L_ijRepresenting the inductance of the transmission line between node i and node j; n is the total number of nodes contained in the line;

step 1.3, setting the Jacobian transfer function matrixes of all the power electronic equipment to be the same under the self-reference capacity, converting the Jacobian transfer function matrixes of all the power electronic equipment to be under the system capacity, and obtaining a dynamic Jacobian transfer function matrix J comprising all the power electronic equipment_Gm(s) and the corresponding transfer functions are:

in the formula, S_B＝diag(S_B1,S_B2,…,S_Bn)，S_BiCapacity of the ith equipment;the notation represents the Kronecker product; g_Pθ(s) transfer function, G, representing the relationship between the active power on the power electronics side and the phase angle of the voltage on the network side_PU(s) a transfer function representing the relationship between the active power on the power electronics side and the voltage amplitude on the network side, G_Qθ(s) a transfer function representing the relationship between the reactive power on the power electronics side and the phase angle of the voltage on the network side, G_QU(s) a transfer function representing the relation between the reactive power of the power electronics device side and the voltage amplitude of the grid side;

the set wind speed is constant, namely the output of the fan is constant, and the direct current capacitance link is as follows:

in the formula,. DELTA.V_dcRepresenting the fluctuation quantity of voltage at two ends of the direct current capacitor; delta P_gRepresenting the fluctuation quantity of the active power at the network side; c represents a dc capacitance value; v_dc0Representing the steady-state value of the voltage at two ends of the direct current capacitor;

substitution of formula (7) into J_GmIn the step(s), the dynamic jacobian transfer function of the power electronic equipment comprising the direct current capacitance link is obtained as follows:

in the formula, G₁(s) is the transfer function of the power outer loop control loop; g₂(s) is the current inner loop control loop transfer function; g_PLL(s) is a transfer function of the phase-locked loop control loop; l is the filter inductance; i is_dIs the net side d-axis current; i is_qIs the net side q-axis current; u shape_dIs a grid-connected voltage;

step 1.4, mixing J_netm(s) and J_GmThe concrete form of(s) is substituted for the closed-loop characteristic equation of the multi-feed system of formula (1) as follows:

in the formula, det (×) represents a determinant;

simultaneous left multiplication matrix of terms in equation (9)

The closed-loop characteristic equation of the system becomes:

in the formula, P_b＝P_i/S_Bi，Q_b＝Q_i/S_Bi；I_n∈R^n×nIs an identity matrix; j. the design is a square_eqTo expand the admittance matrix, the expression is:

extended admittance matrix J_eqThe eigenvalue decomposition can be carried out, all eigenvalues are positive numbers, and the algebraic multiplicity and geometric multiplicity of the minimum eigenvalue are both 1, so that the existence of the invertible matrix W can be applied to J_eqCarrying out similar diagonal transformation to satisfy

W^-1J_eqW＝Λ＝-diag(λ_i) (12)

In the formula, Λ represents J_eqIs a characteristic value of_iA diagonal matrix arranged in order; diag denotes determinant;

step 1.5, because G in formula (10)_Pθ(s)I_n、G_PU(s)I_n、G_Qθ(s)I_n、G_QU(s)I_n、diag(P_b) And diag (Q)_b) All are diagonal matrices, so formula (12) is substituted for formula (10):

at this time, each term in the system closed-loop characteristic equation (8) is a diagonal matrix, so equation (8) is rewritten as follows:

in the formula (I), the compound is shown in the specification,

represents determinant multiplication;

namely, it is

c₁(s)×c₂(s)…×c_n(s)＝0 (15)

In the formula, c_i(s) a closed loop equation representing the ith branch of the system;

wherein

Each factor c in the closed-loop characteristic equation of the multi-feed system_iAnd(s) has the same form as the closed-loop characteristic equation of the single-feed system, so that the closed-loop characteristic equation of the n-feed system is the product of the closed-loop characteristic equations of the n single-feed systems.

Further, the capacitance voltage is expressed by a sine function in step 2, and the direct-current capacitance oscillation frequency capable of exciting the network-side low-frequency oscillation mode is obtained by a frequency sweep method, which specifically includes:

step 2.1, mixing V_dcSince the transfer function is represented by the frequency domain, the fluctuation of (2) is represented by a sine function, which is subjected to a laplacian transform to obtain:

in the formula, ω₁Represents the oscillation frequency of the direct current capacitor;

step 2.2, substituting formula (17) into J_GmIn(s), the jacobian transfer function for obtaining the dynamic state of the power electronic device including the capacitance fluctuation is as follows:

step 2.3, comparing the characteristic roots without fluctuation and with fluctuation by a frequency sweeping method to obtain the direct current capacitance oscillation frequency omega capable of exciting the network side low-frequency oscillation mode₁。

Further, the establishing of the wind power grid-connected machine side model in the step 3 obtains a transfer function of the direct current capacitor with respect to the rotating speed of the fan, and specifically includes the following steps:

step 3.1, setting the power of the network side to be free of fluctuation, and establishing a direct current capacitance model:

in the formula,. DELTA.V_dcRepresenting voltage fluctuation values on two sides of the direct current capacitor; delta P_wRepresenting the machine side active power fluctuation value;

step 3.2, establishing a machine side model, wherein the Jacobian transfer function matrix from the fan to the direct current capacitor is as follows:

in the formula, v_ds0、v_qs0Steady-state voltages of a d axis and a q axis of the stator are respectively; i.e. i_ds0、i_qs0The steady-state currents are respectively a d axis and a q axis of the stator; j. the design is a square_MIs the rotational inertia of the fan; p is the number of pole pairs of the fan; psi is the flux linkage value of the permanent magnet fan; delta omega is a machine side rotating speed fluctuation value; a represents machine side active power fluctuation and i_ds0A value of interest; b represents machine side active power fluctuation and v_ds0A value of interest; c represents machine side active power fluctuation and i_qs0A value of interest;

and 3.3, substituting the direct current capacitor model into a formula (20) to obtain a transfer function of the rotating speed to the voltage at two ends of the direct current capacitor, wherein the transfer function is as follows:

further, the step 4 of establishing a model of influence of wind speed on the rotating speed of the fan, and analyzing to obtain a wind speed fluctuation frequency band capable of exciting the direct current capacitance oscillation frequency in the step 2, specifically as follows:

step 4.1, wind speed v_windFor the input variable, the relation between the output power of the fan and the wind speed is obtained as follows:

wherein ρ is an air density; pi is the circumference ratio; r is the length of the fan blade; v. of_windIs the wind speed; p_mOutputting power for the fan;

and 4.2, expressing the wind speed fluctuation by a sine function:

in the formula, v_wind0Is a steady state value of wind speed; omega_windRepresenting the wind speed oscillation frequency;

the shafting model of the generator related to the wind speed is as follows:

in the formula, J_wRepresenting the generator moment of inertia; t is_mRepresenting generator mechanical torque; t is_eRepresenting the electromagnetic torque of the generator; omega_m0Representing a steady state value of the rotating speed of the generator rotor;

4.3, linearizing the formula (24) to obtain the influence of the wind speed on the rotating speed, and obtaining a wind speed fluctuation frequency band capable of exciting low-frequency oscillation of the direct current capacitor voltage by using a frequency sweeping method;

and 4.4, analyzing the influence of each controller parameter on the oscillation mode to obtain a dominant oscillation mode, and verifying the small signal.

Further, the coordinating and optimizing parameters of the machine side controller and the network side controller by using the particle swarm optimization in the step 5 enables the corresponding modal damping ratio to be maximum, so as to suppress low-frequency oscillation, specifically as follows:

step 5.1, controlling the proportional parameter PLL of the loop by the phase-locked loop_lAnd integral parameter PLL_iThe method comprises the following steps of establishing a multi-controller parameter coordination optimization model by taking a power outer ring proportional parameter kp1, a power outer ring integral parameter ki1, a stator d-axis current inner ring proportional parameter kp2, a stator d-axis current inner ring integral parameter ki2, a stator q-axis current inner ring proportional parameter kp3 and a stator q-axis current inner ring integral parameter ki3 of a machine-side controller, taking a grid-side d-axis current inner ring proportional parameter gp1, a power outer ring integral parameter gi1, a grid-side d-axis current inner ring proportional parameter gp2, a grid-side d-axis current inner ring integral parameter gi2, a grid-side q-axis current inner ring proportional parameter gp3 and a grid-side q-axis current inner ring integral parameter gi3 of a grid-side controller as optimization variables, taking an oscillation mode caused by wind speed fluctuation as a target mode, and taking a target mode damping ratio maximization as an optimization target:

in the formula, η_iDetermining the weight coefficient of the target mode i by a dispatching department according to the actual condition of the power system so as to reflect the attention degree of the dispatching department to different target modes; zeta_iDamping ratio for target mode i; k is the target mode number; f is the maximum system damping ratio of the target mode; max () represents taking the maximum value; Σ denotes a summation function;

step 5.2, in the process of suppressing oscillation by coordinating and optimizing parameters of the network side controller and the machine side controller, the following constraint conditions need to be met: (1) the damping ratio of the oscillation mode caused by wind speed fluctuation needs to be larger than a set threshold value rho₀(ii) a (2) The damping ratio of other oscillation modes which do not need to be optimized is not lower than the set threshold value rho₁Meanwhile, when the oscillation mode damping ratio which does not need to be optimized falls, the falling amplitude needs to be within the lower limit value constraint; (3) the controller parameters to be optimized need to satisfy their own parameter constraints;

then the constraint conditions for obtaining the coordination optimization are as follows:

in the formula, k_p-x,yThe y scale factor of the x controller; k is a radical of_p-x,y-maxAnd k is_p-x,y-minAre each k_p-x,yThe corresponding upper and lower limits; k is a radical of_i-x,yThe ith integral coefficient of the xth controller; k is a radical of_i-x,y-maxAnd k is_i-x,y-minAre each k_i-x,yThe corresponding upper and lower limits; rho_0i1A damping ratio threshold for a given target mode i; rho_0i2ξ as the damping ratio of the target mode i under the initial parameters of the controller_kDamping ratio for non-target mode k, Δ ξ_kThe damping ratio variation before and after coordination optimization for the non-target mode k; rho_1kA damping ratio threshold for a given non-target mode k; rho_2kPercent change in damping ratio for a given non-target mode k, which is a positive number;

and 5.3, obtaining the dominant controller parameters of the target oscillation mode in the step 4.4, solving different dominant controller parameters by using a particle swarm algorithm, obtaining the controller parameter value corresponding to the maximum target function shown in the formula (25), namely the controller parameter value of the mode with the maximum damping ratio, and accordingly inhibiting oscillation of the mode.

Compared with the prior art, the invention has the remarkable advantages that: (1) the method has the advantages of a generalized short-circuit ratio method and a small-signal analysis method, not only simplifies the circuit, but also comprehensively considers the influence of the frequency of the wind speed on the small interference stability of the whole wind power grid-connected system, analyzes the influence of the wind speed on the wind power system in detail from the mechanism, and has guiding significance for researching the problem analysis of wind speed fluctuation and forced oscillation generated by the system; (2) aiming at the problems of wind speed fluctuation and forced oscillation generated by a system, a damping controller and other suppression devices are not connected, so that the hardware cost is saved; (3) the wind power grid-connected low-frequency oscillation is suppressed through the joint debugging machine side controller and the grid side controller, and the influence of the wind speed fluctuation frequency is analyzed from the mechanism, so that the parameter optimization dimensionality of the controller is simplified, and the suppression step of the low-frequency oscillation caused by the wind speed is greatly simplified.

Drawings

FIG. 1 is a schematic flow diagram of a wind power integration low-frequency oscillation suppression method based on a generalized short-circuit ratio method.

Fig. 2 is a circuit configuration diagram of the generalized short-circuit ratio method multi-feed system of the present invention.

Fig. 3 is a diagram illustrating simulation results of network-side line simplification by the generalized short-circuit ratio method according to an embodiment of the present invention.

FIG. 4 is a graph of amplitude-frequency characteristics of a system for forced oscillation of a wind speed frequency excitation system in an embodiment of the invention.

Detailed Description

The invention is described in further detail below with reference to the figures and the embodiments.

With reference to fig. 1, the wind power integration low-frequency oscillation suppression method based on the generalized short-circuit ratio method includes the following steps:

step 1, establishing a wind power grid-connected machine side model by using a generalized short-circuit ratio method, and simplifying a circuit by using the generalized short-circuit ratio method according to a direct current capacitance model, wherein the generalized short-circuit ratio method specifically comprises the following steps:

step 1.1, combining with the graph 2, dividing a machine side model in wind power integration into a Jacobian transfer function matrix J of a power electronic device group_Gm(s) and Jacobian transfer function matrix J for an AC power grid_netm(s)：

Wherein, Δ P represents the fluctuation value of the active power of the network side, Δ Q represents the fluctuation value of the reactive power of the network side, Δ θ represents the fluctuation value of the phase angle of the voltage of the network side, Δ U represents the fluctuation value of the voltage of the network side, and U represents the amplitude of the voltage of the network side;

step 1.2, Jacobian transfer function matrix J of alternating current power grid_netm(s) is:

wherein, J_Pθ(s) a transfer function representing the relationship between the active power of the AC network and the phase angle of the network-side voltage, J_PU(s) a transfer function representing the relationship between the active power of the AC network and the network-side voltage amplitude, J_Qθ(s) representing the reactive power of the AC mainsTransfer function related to phase angle of grid side voltage, J_QU(s) a transfer function representing the relation between the reactive power of the ac power grid and the grid side voltage amplitude;

in the formula (I), the compound is shown in the specification,

wherein the content of the first and second substances,

wherein ω is₀The power frequency angular velocity; p_i＝diag(PL₁,PL₂,…,PL_n)，P_iIs a per unit value, PL, of the output power of the power electronic equipment based on unit capacity₁,PL₂,…,PL_nIndicating the power of the grid-connected line injection node 1, 2, …, n; m is a cosine value of the current injected between the nodes i and j; n is the sine value of the injected current between the nodes i and j;

wherein:

in the formula of U_i、U_jRepresents the node voltages of the node i and the node j; theta_ijRepresents the voltage phase angle difference between node i and node j;

B_ijin order to eliminate elements in the node admittance matrix B after the nodes connected with the infinite power grid, the specific expression is as follows:

in the formula, L_ijRepresenting the inductance of the transmission line between node i and node j; n is the total number of nodes contained in the line;

step 1.3, setting the Jacobian transfer function matrixes of all the power electronic equipment to be the same under the self-reference capacity, converting the Jacobian transfer function matrixes of all the power electronic equipment to be under the system capacity, and obtaining the dynamic Jacobian transfer function containing all the power electronic equipmentNumber matrix J_Gm(s) and the corresponding transfer functions are:

in the formula, S_B＝diag(S_B1,S_B2,…,S_Bn)，S_BiCapacity of the ith equipment;

the notation represents the Kronecker product; g_Pθ(s) transfer function, G, representing the relationship between the active power on the power electronics side and the phase angle of the voltage on the network side_PU(s) a transfer function representing the relationship between the active power on the power electronics side and the voltage amplitude on the network side, G_Qθ(s) a transfer function representing the relationship between the reactive power on the power electronics side and the phase angle of the voltage on the network side, G_QU(s) a transfer function representing the relation between the reactive power of the power electronics device side and the voltage amplitude of the grid side;

the set wind speed is constant, namely the output of the fan is constant, and the direct current capacitance link is as follows:

in the formula,. DELTA.V_dcRepresenting the fluctuation quantity of voltage at two ends of the direct current capacitor; delta P_gRepresenting the fluctuation quantity of the active power at the network side; c represents a dc capacitance value; v_dc0Representing the steady-state value of the voltage at two ends of the direct current capacitor;

substitution of formula (7) into J_GmIn the step(s), the dynamic jacobian transfer function of the power electronic equipment comprising the direct current capacitance link is obtained as follows:

in the formula, G₁(s) is the transfer function of the power outer loop control loop; g₂(s) is the current inner loop control loop transfer function; g_PLL(s) is a transfer function of the phase-locked loop control loop; l is the filter inductance;I_dis the net side d-axis current; i is_qIs the net side q-axis current; u shape_dIs a grid-connected voltage;

step 1.4, mixing J_netm(s) and J_GmThe concrete form of(s) is substituted for the closed-loop characteristic equation of the multi-feed system of formula (1) as follows:

in the formula, det (×) represents a determinant;

simultaneous left multiplication matrix of terms in equation (9)

The closed-loop characteristic equation of the system becomes:

in the formula, P_b＝P_i/S_Bi，Q_b＝Q_i/S_Bi；I_n∈R^n×nIs an identity matrix; j. the design is a square_eqTo expand the admittance matrix, the expression is:

extended admittance matrix J_eqEigenvalue decomposition can be carried out, all eigenvalues are positive numbers, algebraic multiplicity and geometric multiplicity of minimum eigenvalues are 1, and therefore reversible matrix W can be used for J_eqCarrying out similar diagonal transformation to satisfy

W^-1J_eqW＝Λ＝-diag(λ_i) (12)

In the formula, Λ represents J_eqIs a characteristic value of_iA diagonal matrix arranged in order; diag denotes determinant;

step 1.5, because G in formula (10)_Pθ(s)I_n、G_PU(s)I_n、G_Qθ(s)I_n、G_QU(s)I_n、diag(P_b) And diag (Q)_b) All are diagonal matrices, so formula (12) is substituted for formula (10): :

at this time, each term in the system closed-loop characteristic equation (8) is a diagonal matrix, so equation (8) is rewritten as follows:

in the formula (I), the compound is shown in the specification,

represents determinant multiplication;

namely, it is

c₁(s)×c₂(s)…×c_n(s)＝0 (15)

In the formula, c_i(s) a closed loop equation representing the ith branch of the system;

wherein

Each factor c in the closed-loop characteristic equation of the multi-feed system_iAnd(s) has the same form as the closed-loop characteristic equation of the single-feed system, so that the closed-loop characteristic equation of the n-feed system is the product of the closed-loop characteristic equations of the n single-feed systems, and the calculation dimension is simplified.

The simulation is carried out on the model, and FIG. 3 is a schematic diagram of the simulation result, so that the multi-feed system can select the equivalent value by using n single-feed systems.

Step 2, representing the capacitor voltage by a sine function, and obtaining the direct current capacitor oscillation frequency capable of exciting the network side low-frequency oscillation mode by a frequency sweeping method, wherein the direct current capacitor oscillation frequency is as follows:

step 2.1, mixing V_dcSince the transfer function is represented by the frequency domain, the fluctuation of (2) is represented by a sine function, which is subjected to a laplacian transform to obtain:

in the formula, ω₁Represents the oscillation frequency of the direct current capacitor;

step 2.2, substituting formula (17) into J_GmIn(s), the jacobian transfer function for obtaining the dynamic state of the power electronic device including the capacitance fluctuation is as follows:

step 2.3, comparing the characteristic roots without fluctuation and with fluctuation by a frequency sweeping method to obtain the direct current capacitance oscillation frequency omega capable of exciting the network side low-frequency oscillation mode₁。

Step 3, establishing a side model of the wind power grid-connected machine to obtain a transfer function of the direct current capacitor about the rotating speed of the fan, wherein the transfer function specifically comprises the following steps:

step 3.1, setting the power of the network side to be free of fluctuation, and establishing a direct current capacitance model:

in the formula,. DELTA.V_dcRepresenting voltage fluctuation values on two sides of the direct current capacitor; delta P_wRepresenting the machine side active power fluctuation value;

step 3.2, establishing a machine side model, wherein the Jacobian transfer function matrix from the fan to the direct current capacitor is as follows:

in the formula, v_ds0、v_qs0Steady-state voltages of a d axis and a q axis of the stator are respectively; i.e. i_ds0、i_qs0The steady-state currents are respectively a d axis and a q axis of the stator; j. the design is a square_MIs the rotational inertia of the fan; p is the number of pole pairs of the fan; psi is the flux linkage value of the permanent magnet fan; delta omega is a machine side rotating speed fluctuation value; a represents machine side active power fluctuation and i_ds0A value of interest; b represents machine side active power waveMotion and v_ds0A value of interest; c represents machine side active power fluctuation and i_qs0A value of interest;

and 3.3, substituting the direct current capacitor model into a formula (20) to obtain a transfer function of the rotating speed to the voltage at two ends of the direct current capacitor, wherein the transfer function is as follows:

step 4, establishing a model of influence of wind speed on the rotating speed of the fan, and analyzing to obtain a wind speed fluctuation frequency band capable of exciting the direct current capacitance oscillation frequency in the step 2, wherein the model specifically comprises the following steps:

step 4.1, wind speed v_windFor the input variable, the relation between the output power of the fan and the wind speed is obtained as follows:

wherein ρ is an air density; pi is the circumference ratio; r is the length of the fan blade; v. of_windIs the wind speed; p_mOutputting power for the fan;

and 4.2, expressing the wind speed fluctuation by a sine function:

in the formula, v_wind0Is a steady state value of wind speed; omega_windRepresenting the wind speed oscillation frequency;

the shafting model of the generator related to the wind speed is as follows:

in the formula, J_wRepresenting the generator moment of inertia; t is_mRepresenting generator mechanical torque; t is_eRepresenting the electromagnetic torque of the generator; omega_m0Representing a steady state value of the rotating speed of the generator rotor;

4.3, linearizing the formula (24) to obtain the influence of the wind speed on the rotating speed, and obtaining a wind speed fluctuation frequency band capable of exciting low-frequency oscillation of the direct current capacitor voltage by using a frequency sweeping method;

and 4.4, analyzing the influence of each controller parameter on the oscillation mode to obtain a dominant oscillation mode, and verifying the small signal.

As can be seen from FIG. 4, when the wind speed fluctuation frequency approaches the oscillation frequency of 1Hz, the amplitude-frequency characteristic curve of the transfer function has an extreme point, i.e. forced oscillation is excited.

Step 5, coordinating and optimizing parameters of the machine side controller and the network side controller by utilizing a particle swarm algorithm to enable the corresponding modal damping ratio to be maximum, so that the effect of inhibiting low-frequency oscillation is achieved, and the method specifically comprises the following steps:

step 5.1, controlling the proportional parameter PLL of the loop by the phase-locked loop_lAnd integral parameter PLL_iThe method comprises the following steps of establishing a multi-controller parameter coordination optimization model by taking a power outer ring proportional parameter kp1, a power outer ring integral parameter ki1, a stator d-axis current inner ring proportional parameter kp2, a stator d-axis current inner ring integral parameter ki2, a stator q-axis current inner ring proportional parameter kp3 and a stator q-axis current inner ring integral parameter ki3 of a machine-side controller, taking a grid-side d-axis current inner ring proportional parameter gp1, a power outer ring integral parameter gi1, a grid-side d-axis current inner ring proportional parameter gp2, a grid-side d-axis current inner ring integral parameter gi2, a grid-side q-axis current inner ring proportional parameter gp3 and a grid-side q-axis current inner ring integral parameter gi3 of a grid-side controller as optimization variables, taking an oscillation mode caused by wind speed fluctuation as a target mode, and taking a target mode damping ratio maximization as an optimization target:

in the formula, η_iDetermining the weight coefficient of the target mode i by a dispatching department according to the actual condition of the power system so as to reflect the attention degree of the dispatching department to different target modes; zeta_iDamping ratio for target mode i; k is the target mode number; f is the maximum system damping ratio of the target mode; max () represents taking the maximum value; Σ denotes a summation function;

step 5.2, the parameters of the controllers on the network side and the machine side are coordinated, optimized and oscillation suppressedIn the process (2), the following constraint conditions need to be satisfied: (1) the damping ratio of the oscillation mode caused by wind speed fluctuation needs to be larger than a set threshold value rho₀(ii) a (2) The damping ratio of other oscillation modes which do not need to be optimized is not lower than the set threshold value rho₁Meanwhile, when the oscillation mode damping ratio which does not need to be optimized falls, the falling amplitude needs to be within the lower limit value constraint; (3) the controller parameters to be optimized need to satisfy their own parameter constraints;

then the constraint conditions for obtaining the coordination optimization are as follows:

in the formula, kp-x, y is the y scale factor of the x controller; kp-x, y-max and kp-x, y-min are the upper and lower limit values corresponding to kp-x, y respectively; ki-x, y is the y integral coefficient of the x controller; ki-x, y-max and ki-x, y-min are respectively the upper limit value and the lower limit value corresponding to ki-x, y; rho_0i1A damping ratio threshold for a given target mode i; rho_0i2ξ as the damping ratio of the target mode i under the initial parameters of the controller_kDamping ratio for non-target mode k, Δ ξ_kThe damping ratio variation before and after coordination optimization for the non-target mode k; rho_1kA damping ratio threshold for a given non-target mode k; rho_2kPercent change in damping ratio for a given non-target mode k, which is a positive number;

and 5.3, obtaining the dominant controller parameters of the target oscillation mode in the step 4.4, solving different dominant controller parameters by using a particle swarm algorithm, obtaining the controller parameter value corresponding to the maximum target function shown in the formula (25), namely the controller parameter value of the mode with the maximum damping ratio, and accordingly inhibiting oscillation of the mode.

In conclusion, the method has the advantages of a generalized short-circuit ratio method and a small-signal analysis method, not only simplifies the circuit, but also comprehensively considers the influence of the frequency of the wind speed on the small interference stability of the whole wind power grid-connected system, analyzes the influence of the wind speed on the wind power system in detail from the mechanism, and has guiding significance for researching the problem analysis of wind speed fluctuation and forced oscillation generated by the system; aiming at the problems of wind speed fluctuation and forced oscillation generated by a system, a damping controller and other suppression devices are not connected, so that the hardware cost is saved; the wind power grid-connected low-frequency oscillation is suppressed through the joint debugging machine side controller and the grid side controller, and the influence of the wind speed fluctuation frequency is analyzed from the mechanism, so that the parameter optimization dimensionality of the controller is simplified, and the suppression step of the low-frequency oscillation caused by the wind speed is greatly simplified.